Delta E Color Tolerance

[bpolymn]

Hello All,

Fairly new to this forum and have been in injection molding for a couple years. I have a customer who at one point agreed to a color Delta E of </= 3.5. Now this customer is coming back with a requirement of a Delta E of <1.5. This is for a decking product that is stand alone and not required to match anything else near it other than board to board where they meet. The customer does not hold any other regulatory statutes nor requires us to hold other than ISO9001. I looked into it and cannot find anywhere that gives an injection molding industry standard that doesn't involve medical, automotive or aerospace. Can anyone give me an idea of what your customers require or where the industry standard is?

Thanks in advance

[chrisprocess]

Great question, I'm not familiar with an "industry standard" for color matching variation..
In the past customer specifies a color, we'll run off a High and Low limit of that color (example 4% at +/- 1% High and Low) which are then approved by the customer. This is our effective processing window.

Color can be very tricky to quantify in injection molding. It changes by means of lighting, wall thickness, and even surface finish. So on one single part alone with different wall thickness and surface finishes can exhibit various iterations of that color even though the material is the same throughout. Not to mention everyone sees color a little bit differently.

The best way to get color approval is to physically send samples to the customer (High, Low, Nominal). If you open can of worms with Delta E variation you'd need a dang good test method and use Gage R&R to verify the test method is accurate and repeatable.
Or you could try to prove one sample alone exhibits color variation because of the customer-specified surface finish/wall thickness variations in itself


Maybe others can chime in

[MNDesign81]

I suspect your customer hired a recently graduated engineering student.
I used to work for a molder specializing in cosmetic packaging. We had a "standard" of a Delta E of <3.
The QC Lab used a Minolta Photo-spectrometer for the testing in a controlled lighting situation [I don't remember the color temp but it was speced]
The appearance of a Delta E differential varies with any specific color.
For example we had a red [Lady Stetson], samples that would qualify under the "standard" were night & day .... tomato red to fire engine red .... Stevie Wonder could tell it was not right!
On the other side, we had a White [Liz Claiborn] with a difference that was imperceptible to the naked eye but had a delta E of almost 7.
My color house used to laugh at me every time I quoted a delta E.
Good luck with that,

[Pilot]

Hi,
I can confirm, that delta E +/- 1,5 is a standard request from various automotive customers.
what MUST be specified also is a measuring device (Minolta, x-rite usualy) Both sides have to have the same device - best variant.
MUST be specified meassuring points on the parts.
You can contact you color masterbatch supplier for more info. They have a lot experiences.
Also you can study the QA requests from automotive - https://www.google.cz/url?sa=t&rct=j...o0WIfl3DBtj50I

[rickbatey]

Your customer has more than likely had a complaint about different production lots not matching one another. But there are several process variables that affect color. High temperature, residence time, and shear heating can shift colors red or yellow. Plus since this is a decking material UV and FR additives can degrade and shift color. One question not asked is what are you using to add the color concentrate? I’ve used many brands of auger feeders and weight blenders; they are NOT all created equally!! Do you use a reputable color supplier? We actually run end caps for an IP out of specific black color to ensure they match the IP skins. But remember if they want to tighten the specs then they will need to pay for the improved color control!
Rick

[bpolymn]

Hey all thanks for the input you have all hit the nail on the head regarding why they are asking and whats going on. We use a highly recognized color supplier that holds a DE of 0.5 on our incoming color which is in liquid form. We do you a photospectrometer unfortunately not one of the brands you recommended. I did some testing and we have massive issues that are staring us in the face. I have complied over 500 boards running different parameters set on a displacement pump and peristaltic pump. We have determined that injection pressure, injection speed, holding pressure, shot size, barrel temps and cycle time are the most prominent conditions that need to be monitored to ensure color accuracy and consistency. Once the cycle has stabilized we record those parameters as well as the LDR, Shot weight and calibration weight of the colorant. We have been able to achieve a DE of 1.2 or lower until someone adjust the press. So we have had to install another quality inspection to monitor shot weight to re-calibrate the pumps to the current part running. On top of that we order our colorant in 30 gallon drums so we are premixing the barrels prior to use and if they rest over the weekend they are mixed prior to use on monday mornings. We have seen some jumps to a DE of 1.8 and 1.9 but they are rare and controlled. I did a lot of research and there is so much conflicting evidence out there stating that a normal human eye cannot detect a DE of 3 and others say 4. Makes sense why automotive is at a 1.5 since you can be at a DE of 1.5 on opposing sides of the spectrum and still have part to part be under a DE of 3.0 from each other instead of from the standard. I contacted a local decking manufacturer and they require under a 2.2 so we are working with the customer to bring up their tolerance to a 2.0. If not their cost will increase and we will need to start sourcing concentrated material and gravimetric hoppers just for this product line.

again appreciate all the input

[chrisprocess]

Have you looked into or tested pre-compounded material?

[bpolymn]

Have it would nearly double the cost that the customer is not willing to absorb. Pre-compounded material was my first thought to control it.

[bpolymn]

MNDesign81 all of our colors are a grey color that vary from 0-4.5 on the a spectrum, 3-9 on the b spectrum and 60-65 on the L. all really tight together but it's just two colors that are drastically different.

[chrisprocess]

Have it would nearly double the cost that the customer is not willing to absorb. Pre-compounded material was my first thought to control it.

Yes more expensive but arguably very precise to color match. And all the colorant mixing data should be readily available from the supplier.
Everything comes at a cost though! A cost the customer isn't likely willing to pay of course

Would be interested to know if color fade or color drift happens over time in regards to shelf life with these parts.
Like matching shingles to a roof. The shingles that sat outside home depot the longest clearly look different and depict color mismatch from shingles fresh from the factory.